skip to content
Logo Joseph Joyal

About Me...

Professional Summary

I'm a Fullstack Developer with close to 3 years of professional experience, focused on Backend Development, AI Agentic Systems Development and Prompt Engineering. I'm Knowledgable in Typescript and Python, having hands-on experience building microservices from scratch using NodeJs(javascript) and FastAPI(python), utilizing event buses such as Kafka and RabbitMQ to build scalable API’s in GraphQL and REST. I possess exposure working with Docker, writing Dockerfiles and in orchestrating services using Docker Compose. I'm knowledgable in both SQL and NoSQL databases such as PostgreSQL and MongoDB, and in ORM such as TypeORM. I'm also familiar with caching tools such as Redis and search tools such as Elasticsearch and OpenSearch. I'm additionally experienced in Prompt Engineering and using LLM models to develop AI agents and tools using effective prompting, RAG models, langchain, etc. I'm capable of contributing effectively to team objectives and delivering high-quality software solutions while adhering to Agile and Scrum methodologies.

Professional Experience

Virtual Mark Consultancy India Pvt. Ltd.

Client: AstraZeneca

Role: Software Engineer | April 2024 - Present | Hyderabad, India (Fully Remote)

Key Responsibilities:

  • Developed multiple POC's on AI Agents exploring various technologies, LLM providers and tools such as Langchain, AWS Bedrock, AzureOpenAI using various LLM models.
  • Worked on multiple Langchain + ReAct based agents with customized tools to answer domain-specific questions integrated with internal client applications.
  • Developed and integrated various agent-tools for the above mentioned project(such as Text-To-API, Text-To-SQL, etc).
  • Utilised AWS API-Gateway for creating websocket endpoints which was utilized to stream responses generated by agent from server to client for the above mentioned projects.
  • Implemented various REST APIs for client-server communication as per requirements, collaborated with frontend teams for API integration.
  • In another project, existing API's for a dashboard were enhanced to support a new pharmaceutical drug, updating the Elasticsearch/OpenSearch queries to accommodate this change as per the requirements.
  • Integrated Google Gemini 2.5 Pro and Flash models into the multi-agent AI project, enabling improved reasoning capabilities and more cost-effective agent usage.

Some Of The Technologies Used:

  • Node.js/Typescript
  • Python/FastAPI
  • PostgreSQL
  • Elasticsearch/Opensearch
  • Langchain
  • Vector Databases
  • AWS Bedrock
  • Google Vertex AI
  • Azure OpenAI
  • Websockets
  • API Gateway
  • REST API's
  • Prompt Engineering
  • Multi-Agentic AI
  • Tool Calling/Function Calling AI Agents
  • Redis (For Caching and Messaging)
Appt Software Technologies Pvt. Ltd.

Role: Associate Software Engineer | March 2023 - March 2024 | Hyderabad, India (On-Site)

Key Responsibilities:

  • Developed new features as per requirements, refactored and maintained existing code for the core application (Low-Code Platform)
  • Designed and optimized database schemas (For POC Applications and client projects).
  • Implemented GraphQL based event driven API microservices, collaborated with frontend team for optimal integration.
  • Worked on a POC for an app used to parse documents to extract unstructured data using Tesseract OCR.
  • Worked on Keycloak authentication for a client project, implemented firebase authentication for another client project.
  • Refactored the Kafka based event flow for the platform product to make it more optimized and performant, improving the speed of code generation by more than 30%.
  • Implemented a MongoDB based API for a client POC.
  • Worked on the backend + prompt engineering for a POC application (chatbot with different medical research papers as knowledgebase) using Azure Open AI + RAG with PG-Vector for a leading indian pharmaceutical company.

Some Of The Technologies Used Overall:

  • Node.js + Typescript
  • PostgreSQL
  • TypeORM
  • WSL
  • Tesseract OCR
  • Firebase
  • Keycloak Authentication
  • MongoDB
  • Minio
  • EJS
  • Azure Open AI/Langchain
  • RAG
  • Docker/Docker-Compose
  • Kafka
  • Event Driven Microservice

Project 1: Low Code Platform (Company Product)

The product is a low code platform that allows users to build full fledged applications (based on typescript and python) with minimal coding. The platform provides a drag and drop interface to build applications and integrates with various external services to provide a seamless experience.

Key Responsibilities:

  • Refactored and optimized the code generation flow for the Low-Code Application to improve the performance and reduce the time taken for code generation.
  • Added new functionalities such as support to push generated code to various SCM (souce code management) platforms such as github, gitlab and bitbucket.
  • Added the object storage capability using minio, compatable with S3 using the presigned url method (same used in different client projects as well).

Project 2: AI Powered Invoice Parser (Client Project)

This product is an invoice reader that helps users to extract data from invoices and store them in a structured format. The product uses text-parser packages and LLM to extract text from the invoice and structure it, uses Minio/S3 to store the invoice pdf files.

Key Responsibilities:

  • Led the initial serverside development for the application.
  • Added support for multitenancy using firebase.
  • Added support for complex filtering and dashboard views using SQL queries.
  • Implemented GraphQL based API's and collaborated with UI team for integration.
  • Implemented an LLM functionality to structure the parsed unstructured data.

Project 3: CRM Tool (Internal Tool)

This product was used to manage client data and interactions by the company sales team to manage leads, clients, and interact with them, scheduling various marketing emails.

Key Responsibilities:

  • Collaborated with senior developers to designing the database schema and the implementation from the requirements.
  • Added functionality to parse emails from CSV and send customised email for each parsed client using nodemailer and with time interval in between using cron job.
  • Additional support to send scheduled follow-up emails.
Sentiero Tech LLP.

Role: Freelance Junior Developer | November 2022 - March 2023 | Kochi, India (Remote)

Key Responsibilities:

  • Developed a responsive website for client project. The website was implemented based on the client requirements and was completed within a span of 2 weeks.

Some Of The Technologies Used:

  • Javascript
  • HTML
  • CSS/Bootstrap

Let's Connect!

Feel free to get in touch with me if you’re interested in collaborating or discussing any potential opportunities. You can reach out to me over my email (mail@jooyal.com), connect with me on LinkedIn (linkedin.com/in/josephjoyal) or GitHub (github.com/jooyal).